Figure 1 A multi-sector single cell atlas of HCC¶
In [2]:
import numpy as np
import pandas as pd
import scanpy.api as sc
import matplotlib.pyplot as plt
sc.settings.verbosity = 3
sc.settings.set_figure_params(dpi=80)# verbosity: errors (0), warnings (1), info (2), hints (3)
sc.logging.print_versions()
adata = sc.read("./data/Liver10X_73k.h5ad")
In [3]:
sc.settings.set_figure_params(dpi=100)
In [4]:
adata
Out[4]:
Plotting¶
Louvain Clusters¶
In [17]:
sc.settings.set_figure_params(dpi=200)
sc.pl.umap(adata, color='louvain', legend_loc= 'on data',
palette=['#FFFF00','#1CE6FF','#FF34FF','#FF4A46','#008941',
'#006FA6','#A30059','#ff7a7a','#7A4900','#0000A6',
'#63FFAC','#B79762','#004D43','#8FB0FF','#997D87',
'#5A0007','#809693','#1B4400','#af60ff','#4FC601',
'#3B5DFF','#4A3B53','#FF2F80','#61615A','#BA0900',
'#6B7900','#00C2A0','#FFAA92','#FF90C9'], size=15)
Patient ID¶
In [11]:
sc.settings.set_figure_params(dpi=120)
sc.pl.umap(adata, color=['patientno'], size=30)
In [19]:
df = adata.obs.groupby(['patientno']).size()
ax = df.plot(kind='bar', figsize=[20,7] , width=0.9, color=adata.uns['patientno_colors'])
plt.ylabel("# of cells")
plt.xlabel("patientno")
plt.grid(False)
#plt.savefig('./figures_Liv73k/patient_no.cells.pdf')
PNC (Peripheral Tumor, Adjacent Normal, Core Tumor)¶
In [10]:
sc.settings.set_figure_params(dpi=120)
sc.pl.umap(adata, color=['PNC'], size=30, palette=['#e6194b', '#f5bb06', '#81007f'])
Myeloid Circle plot¶
In [23]:
adata_mye = adata[adata.obs.loc[adata.obs.louvain.isin(['8','10'])].index.tolist(), ]
sc.settings.set_figure_params(dpi=100)
sc.pl.umap(adata_mye, color=['louvain','PNC'], size = 25)#, save='_louvain.png')
display(adata_mye)
fig_size = plt.rcParams["figure.figsize"]
fig_size[0] = 6
fig_size[1] = 4
plt.rcParams["figure.figsize"] = fig_size
count = adata_mye.obs.groupby('PNC').size()
df = adata_mye.obs.groupby(['PNC','louvain']).size().unstack().fillna(0)
df = df.div(count, axis=0)
df1 = df.apply(lambda x: x/sum(x)*100, axis = 0).transpose()
ax = df1.plot.barh(stacked=True, color=['#e6194b', '#f5bb06', '#81007f'], width=0.9)
ax.grid(False)
ax.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)
fig = ax.get_figure()
#fig.savefig('./paper_figures1/tcells/barplot_tcells_NT.pdf')
NT = adata_mye.obs.groupby(['louvain','PNC']).size()
NT
fig, ax = plt.subplots()
size = 0.7
vals = np.array(NT.unstack())
outer_colors = adata_mye.uns['louvain_colors']
inner_colors = ['#e6194b', '#f5bb06', '#81007f']
ax.pie(vals.sum(axis=1), radius=2.5-0.25, colors=outer_colors,
wedgeprops=dict(width=size, edgecolor='w'))
ax.pie(vals.flatten(), radius=2.3-0.6, colors=inner_colors,
wedgeprops=dict(width=size, edgecolor='w'))
#ax.set(aspect="equal", title='Pie plot with `ax.pie`')
plt.show()
#fig.savefig('piechart__HCC_mye_PNC.pdf', bbox_inches='tight')
plt.close(fig)
Lymphoid Circle plot¶
In [24]:
adata_lymph = adata[adata.obs.loc[adata.obs.louvain.isin(['0','1','3','4','5','9',
'12','15','22','23'])].index.tolist(), ]
sc.settings.set_figure_params(dpi=100)
sc.pl.umap(adata_lymph, color=['louvain','PNC'], size = 25)#, save='_louvain.png')
display(adata_lymph)
fig_size = plt.rcParams["figure.figsize"]
fig_size[0] = 6
fig_size[1] = 4
plt.rcParams["figure.figsize"] = fig_size
count = adata_lymph.obs.groupby('PNC').size()
df = adata_lymph.obs.groupby(['PNC','louvain']).size().unstack().fillna(0)
df = df.div(count, axis=0)
df1 = df.apply(lambda x: x/sum(x)*100, axis = 0).transpose()
ax = df1.plot.barh(stacked=True, color=['#e6194b', '#f5bb06', '#81007f'], width=0.9)
ax.grid(False)
ax.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)
#fig = ax.get_figure()
#fig.savefig('./paper_figures1/tcells/barplot_tcells_NT.pdf')
NT = adata_lymph.obs.groupby(['louvain','PNC']).size()
NT
fig, ax = plt.subplots()
size = 0.7
vals = np.array(NT.unstack())
outer_colors = adata_lymph.uns['louvain_colors']
inner_colors = ['#e6194b', '#f5bb06', '#81007f']
ax.pie(vals.sum(axis=1), radius=2.5-0.25, colors=outer_colors,
wedgeprops=dict(width=size, edgecolor='w'))
ax.pie(vals.flatten(), radius=2.3-0.6, colors=inner_colors,
wedgeprops=dict(width=size, edgecolor='w'))
#ax.set(aspect="equal", title='Pie plot with `ax.pie`')
plt.show()
#fig.savefig('piechart__HCC_imm_PNC.pdf', bbox_inches='tight')
#plt.close(fig)
Endothelial Circle plot¶
In [25]:
adata_endo = adata[adata.obs.loc[adata.obs.louvain.isin(['2','6','17','26','27'])].index.tolist(), ]
sc.settings.set_figure_params(dpi=100)
sc.pl.umap(adata_endo, color=['louvain','PNC'], size = 25)#, save='_louvain.png')
display(adata_endo)
fig_size = plt.rcParams["figure.figsize"]
fig_size[0] = 6
fig_size[1] = 4
plt.rcParams["figure.figsize"] = fig_size
count = adata_endo.obs.groupby('PNC').size()
df = adata_endo.obs.groupby(['PNC','louvain']).size().unstack().fillna(0)
df = df.div(count, axis=0)
df1 = df.apply(lambda x: x/sum(x)*100, axis = 0).transpose()
ax = df1.plot.barh(stacked=True, color=['#e6194b', '#f5bb06', '#81007f'], width=0.9)
ax.grid(False)
ax.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)
fig = ax.get_figure()
#fig.savefig('./paper_figures1/tcells/barplot_tcells_NT.pdf')
NT = adata_endo.obs.groupby(['louvain','PNC']).size()
NT
fig, ax = plt.subplots()
size = 0.7
vals = np.array(NT.unstack())
outer_colors = adata_endo.uns['louvain_colors']
inner_colors = ['#e6194b', '#f5bb06', '#81007f']
ax.pie(vals.sum(axis=1), radius=2.5-0.25, colors=outer_colors,
wedgeprops=dict(width=size, edgecolor='w'))
ax.pie(vals.flatten(), radius=2.3-0.6, colors=inner_colors,
wedgeprops=dict(width=size, edgecolor='w'))
#ax.set(aspect="equal", title='Pie plot with `ax.pie`')
plt.show()
#fig.savefig('piechart__HCC_endo_PNC.pdf', bbox_inches='tight')
#plt.close(fig)
Viral Status¶
In [17]:
sc.pl.umap(adata, color=['ViralvsNonViral'], size=30)
Expression plot¶
Stroma¶
In [15]:
sc.pl.umap(adata, color=['PECAM1','ACTA2'], size=15)
Epithelial¶
In [16]:
sc.pl.umap(adata, color=['KRT19','ALB'], size=15)
Immune¶
In [18]:
sc.pl.umap(adata, color=['IL7R','CD8A','GNLY','NKG7','MZB1','CD14'], size=15,ncols=3)
Expression Dotplot¶
In [27]:
genes = ['IL7R', 'GNLY', 'SPRY1', 'PLVAP', 'CD8A', 'CD8B', 'CAV1', 'SPINK1', 'ALB', 'LYZ', 'CD7', 'NKG7', 'C1QA',
'CD68', 'APOC3', 'FXYD1', 'TIGIT', 'BATF', 'MYL9', 'ACTA2', 'SERPINA1', 'CYP2E1', 'PRF1', 'CYP2A7', 'CYP4F3',
'SPARC', 'PLVAP', 'MZB1', 'CD79A', 'CD74', 'CD83', 'MLXIPL', 'ORM1', 'TTR', 'STMN1', 'HMGB2', 'KRT18', 'KRT19',
'C3', 'HPX', 'FCN3', 'AKAP12', 'TFF3', 'FABP4', 'KIT']
sc.pl.dotplot(adata, genes, groupby='louvain', dot_max=0.8, dot_min=0.2, standard_scale='var',
color_map='Reds')#, save='_HCC_dotplot.pdf')
Out[27]:
HCC Garnett¶
In [15]:
HCCdf = pd.read_csv('./data/HCCcelltype_marker2.csv',index_col=0)
display(HCCdf.head())
print(HCCdf.shape)
print(set(HCCdf.celltype))
print(len(set(HCCdf.celltype)))
In [16]:
import seaborn as sns
import matplotlib
import matplotlib.pyplot as plt
cmap = matplotlib.cm.ocean.reversed()
def make_cmap(colors, position=None, bit=False):
'''
make_cmap takes a list of tuples which contain RGB values. The RGB
values may either be in 8-bit [0 to 255] (in which bit must be set to
True when called) or arithmetic [0 to 1] (default). make_cmap returns
a cmap with equally spaced colors.
Arrange your tuples so that the first color is the lowest value for the
colorbar and the last is the highest.
position contains values from 0 to 1 to dictate the location of each color.
'''
import matplotlib as mpl
import numpy as np
bit_rgb = np.linspace(0,1,256)
if position == None:
position = np.linspace(0,1,len(colors))
else:
if len(position) != len(colors):
sys.exit("position length must be the same as colors")
elif position[0] != 0 or position[-1] != 1:
sys.exit("position must start with 0 and end with 1")
if bit:
for i in range(len(colors)):
colors[i] = (bit_rgb[colors[i][0]],
bit_rgb[colors[i][1]],
bit_rgb[colors[i][2]])
cdict = {'red':[], 'green':[], 'blue':[]}
for pos, color in zip(position, colors):
cdict['red'].append((pos, color[0], color[0]))
cdict['green'].append((pos, color[1], color[1]))
cdict['blue'].append((pos, color[2], color[2]))
cmap = mpl.colors.LinearSegmentedColormap('my_colormap',cdict,256)
return cmap
fig = plt.figure()
ax = fig.add_subplot(311)
### Create a list of RGB tuples
colors = [(239, 248, 249), (252, 78, 78),(96, 33, 104),(0,0,0)] # This example uses the 8-bit RGB
### Call the function make_cmap which returns your colormap
my_cmap = make_cmap(colors, bit=True)
### Use your colormap
plt.pcolor(np.random.rand(25,50), cmap=my_cmap)
plt.colorbar()
def sns_heatmap(df,title,subtitle):
mat = pd.DataFrame()
for i in range(df.shape[1]):
if i==df.shape[1]-1:
break
for j in range(df.shape[0]):
a = []
for k in range(df.shape[0]):
if j == k:
a.append(df.iloc[j,i]*df.iloc[k,i+1])
else :
a.append(0)
mat[j]=a
mat.columns = df.index
mat.index = df.index
mat = mat.rename_axis(df.columns[1])
mat = mat.rename_axis(df.columns[0], axis="columns")
f, ax = plt.subplots()
sns.heatmap(mat,vmax=1500, vmin=-1, cmap=my_cmap)#
plt.title(title, fontsize=20)
plt.suptitle(subtitle, fontsize=12)
ax.xaxis.set_tick_params(labelsize=8)
ax.yaxis.set_tick_params(labelsize=8)
In [17]:
fig_size = plt.rcParams["figure.figsize"]
# Prints: [8.0, 6.0]
print ("Previous size:", fig_size)
# Set figure width to 12 and height to 9
fig_size[0] = 12
fig_size[1] = 10
plt.rcParams["figure.figsize"] = fig_size
print ("Current size:", fig_size)
HCCdf2 = HCCdf1.drop(['1 lymph 1 TC','2 lymph 1 NKC','3 mye 1','5 endo','6 Hepa', 'Unknown'], axis=1)#
#HCCdf2 = HCCdf2.drop(['1 lymph STMN1'], axis=0)
HCCdf2 = HCCdf2.apply(lambda x : x/sum(x)*100,1).round(0).astype(int)
sns_plot = sns.heatmap(HCCdf2)#, cmap="YlGnBu", linewidths=.009)#, vmax = 500,, annot=True, fmt="d", linewidths=.01
plt.title('\nHCC cell type agnostic correlation\n', fontsize=20)
#fig = sns_plot.get_figure()
#fig.savefig('./finalfig/HCC.png')
Out[17]:
In [ ]: